Short‑term analysis of employment is carried out, on the one hand, using Employment Estimates, a statistical source based on administrative data supplied by employers, particularly social declarations. These data enable precise measurement of changes in employment from quarter to quarter. On the other hand, the Labour Force Survey (LFS) – which is conducted on a sample of individuals – compares employment and unemployment as defined by the International Labour Organization (ILO), and is therefore the only source that can be used to calculate the unemployment rate. It also provides detailed information on labour market integration according to socio‑demographic characteristics.

Following the health crisis in 2020, the sharp rise in employment was accompanied by a greater divergence than in the past between these two sources. From 2020 to 2023, employment increased by 1.4 million jobs in the LFS, but even more in the Employment Estimates, with 338,000 additional job creations. Three main factors explained this divergence: a marked dynamism of employment for types of jobs being more often under‑reported by respondents in the LFS; differences linked to the LFS's scope and sampling; and under‑representation of people born abroad in the LFS.

In 2024, as a result of the slowdown in the labour market, these factors did not contribute to the same extent any longer, and employment as measured by the LFS was slightly more dynamic than that measured by administrative sources.